Review on Machine Learning for Molecular Excited States
In our Review “Molecular excited states through a machine learning lens” in Nature Reviews Chemistry, we provide insights and highlight challenges of the rapidly growing field of machine learning for excited-states simulation and analysis.
Machine learning emerges as a breakthrough technique for improving accuracy and speed of quantum chemical approaches for excited-state simulations such as prediction of excitation energies, electronic spectra, dynamics simulations, and optoelectronic materials design. Machine learning is also capable of completely bypassing quantum chemical methods in simulations, e.g., by learning directly from the experimental data. Analysis of experimental observations with machine learning can enable efficient structure determination and discover rules for materials design.
Optoelectronic materials design is enjoying the developments in machine learning as it not just provides rules for design, but allows for accelerated high-throughput screening of existing databases and can guide the automatic search in the chemical space. Such approaches have already resulted in discovery of novel materials and in the future, we expect to see more examples of intelligent materials discovery, particularly when machine learning is merged with robotic laboratories.
It should be admitted that so far, most of the surveyed studies are in the proof-of-principle stage and lots of research is still needed to make machine learning a useful tool for routine excited-state simulations and analysis. However, more and more studies emerge applying machine learning to solving practical problems supporting our closing statement in the Review that “ML has a bright future in the field of excited-state research.”
Our contributions to the field is ongoing development of methods for nonadiabatic excited-state dynamics and UV/vis absorption spectra simulations.
The Review is dedicated to 100th anniversary of Xiamen University.
- Pavlo O. Dral*, Mario Barbatti*, Molecular excited states through a machine learning lens, Nat. Rev. Chem. 2021, 5, 388–405. DOI: 10.1038/s41570-021-00278-1.
P.S. See also the blog post by Mario Barbatti.
P.P.S. You can also read an excellent review by Julia Westermayr and Philipp Marquetand, which is complementary to our Review.
P.P.P.S. The review highlighted on the homepage of the College of Chemistry and Chemical Engineering at Xiamen University: 机器学习与激发态研究综述.
I would like to join